CHANGE OR HOMEOSTASIS? A SYSTEMS THEORY APPROACH TO DEPRESSION

 

 

Summary

Looking at the onset of depressive states in linear terms, there is often a  problem of distinguishing cause and effect:  did the adverse life events  cause the depression, or did the depression cause the adverse life events?   If we abandon linear thinking and look at depression in systemic terms, the  problem of cause and effect disappears, but it is replaced by another  problem:  is the patient/environment system characterised by homeostasis or  change?  Some depressed patients seem to be spiralling down towards  disaster. Others seem to be stuck in a rut. In systems terms, can  depression be at the same time an agent of change and an agent of stasis  (or even homeostasis)?  The paradox can be resolved if we postulate that  the function of depression is to reconcile the individual to a subordinate  social role;  depression which reconciles to a pre-existing subordinate  position has static properties;  depression which mediates a switch to a  subordinate position from a previously dominant position has properties of  systemic change.

 

 

In formulating a depressive state, it is sometimes difficult to unravel  cause and effect. Is this man depressed because he lost his job, or did he  lose his job because he was depressed?  Is this woman depressed because her  husband left her, or did her husband leave her because she was depressed  and therefore unrewarding?  Some of the causation of depression can be  attributed to life events (Paykel, 1978) or social factors (Brown, 1989),  but the larger part of the causation remains unaccountable for in these  terms.

 

   One way out of this difficulty is to adopt a systemic epistemology, in  which linear relations of cause and effect are less important than  interactive processes. Attempts to apply systems theory to psychiatry have  been sporadic (Wender, 1968;  Gray et al., 1969; Senay, 1973; Marmor, 1983)  except for systemic family therapy which is explicitly based on systems  theory (Hoffman, 1981). Outcomes such as depression are seen as the  endpoints of positive feedback processes (Feldman, 1976), in which small  random deviations from a steady state are amplified progressively until the  system changes and an "event" or a "condition" is said to occur. Alternatively, random and other deviations from the steady state may be  damped by buffering or by negative feedback, so that an outcome in the form  of a change in the system is avoided (Gray et al., 1969).

 

Positive feedback loops in depression

 

At the intrapsychic level, depressed people have a negative view of the  world and themselves (Beck, 1976), which is in turn depressing (Teasdale,  1985). They selectively recall unfavourable events from the past. They  tend to attribute aversive life events to internal, stable, global causes;   and they attribute success to external, unstable or specific causes. They  blame themselves when things go wrong, but they do not take the credit when  things go right. Thus the attributional style which results from the  depression leads to more depression ((Brewin, 1988). Along similar lines,  Pyszczynski and Greenberg (1987) found that depressed patients focussed on  themselves after a negative outcome but avoided self-focus after a positive  outcome, so intensifying failure and minimising success.

   At the musculo-skeletal level, depression may be associated with loss of  poise and disorders of both posture and gait (Sloman et al., 1982) of which  the subjective experience may be depressing and which may lead to aversive 

reactions from others. There may also be pain, which is not only  depressing in itself but may also give rise to ideas of serious malfunction  and disease. There may be a sense of exhaustion, and easy fatiguability of  muscles, leading to beliefs about chronic viral infection. In general, the  body does not seem to be working well in depression, and this reduces the  patient's confidence in his ability to ward off aversive environmental  events.

   At the executive level, the loss of energy associated with depression  leads to failure to carry out tasks and thus to the accumulation of  unfinished business such as unanswered (and even unopened) letters;  so  that increasing evidence of incompetence and failure surround the depressed  person and cannot but have a further depressing effect. Moreover, neglect  of self-care leads to a deterioration of skin, hair and clothes so that to  look in the mirror is a depressing experience. Maruyama (1963) pointed out  the possibility of "mutual amplification....between loss of self-confidence  and poor performance in a neurotic person."

   Even when tasks are carried out, they may be done badly in depression. A mother is unable to give constructive attention to her children but  rather screams at them or even hits them, and so feels herself to be a bad  mother;  and being unable to respond sexually to her husband she feels  herself to be a bad wife;  and these experiences of failure deepen her  depressed mood.

   At the social level, depression leads to isolation which for most people  is depressing. If depressed people avoided those who make them feel bad,  such as enemies, the depression might serve a homeostatic function in this  respect;  but, on the contrary, they avoid friends and relatives who would,  if they were permitted, cheer them up. Even if relatives are not avoided,  the interaction may not be therapeutic:  "...patients may use depressive 

symptoms to elicit sympathy and care from their families, which in turn 

reinforces the maladaptive behaviour and establishes a vicious circle"  (Veiel & Kuhner, 1990).

     Klerman (1974) investigated the "communication of distress" function  of depression, postulating that it served as a cry for help which mobilised  social resources. However, after studying a group of 40 depressed women  and a matched control group he came to the conclusion that the depression  had alienated the women not only from friends and relatives but even from  their husbands and children. Henderson (1974) put forward the idea that   depression was a form of care-eliciting behaviour, particularly when it was  associated with attempted suicide. While this may be true over the short  term, it is the general experience that others avoid depressed people and  do not give them care. They may justify this neglect by categorising them  as lazy or rude. As the old saying goes, "Laugh, and the world laughs with  you;  weep, and you weep alone."  In more behavioural terms, "Depressed  people are usually not reinforcing to be with and consequently are often  tactfully avoided" (MacLean, 1976, p.313).

   At the therapeutic level, the depressed person tries to get himself out  of the trough, even without the added stimulus from friends to "pull  yourself together" or to "snap out of it."  These efforts fail, and this  failure enhances the depression.

   Criticism of a person because of their depressive symptoms is an  additional positive feedback loop. The criticism makes them more  depressed, which increases the symptoms, which increases the criticism, and  so on. A number of symptoms may be targets for attack, paticularly in the  case of a depressed spouse. A husband may get angry because the house  isn't cleaned properly. Some husbands get angry at lack of sexual  response, others at weight loss or gain. Wives get angry with the husband  who doesn't do jobs around the house, or leaves them half done.

 

   In summary, many authors have commented on the occurrence of positive 

feedback loops in depression, such that a lowering of mood causes changes  in the environment which in turn cause a further lowering of mood.

 

Depression associated with homeostasis

 

In view of the above considerations we might expect every depressed patient  to be accelerating towards disaster, but in practice the majority of  depressed patients, certainly most of those seen in the out-patient clinic,  seem to be very chronic and appear to the clinician to be "stuck" rather  than in a state of change. A man is depressed in his job, but he lacks the  intitiative to apply for another job;  he is nervous about the interview  situation, and he dreads the rejection of being turned down. As with a  job, so with a marriage. A common presentation in the out-patient clinic  is the woman who is married to an uncaring tyrant;  her life is one of  drudgery and service to a man who will give her no pleasure and denies her  the opportunity of seeking pleasure elsewhere. These women are "stuck" in  their marriages, and the depression makes it impossible for them to leave. They lack the energy and initiative to set up on their own, they lack the  interest to look for an alternative partner and the depression makes them  unattractive to any man who might come along.

   The theorist who has most clearly made the case for depression as an  agent of stasis, if not of homeostasis, is Costello (1976), who sees the  function of depression as maintaining the status quo. If mood is  pervasive, depression lowers all incentives equally, so that a failing  incentive is less likely to be replaced by another. 

   In the family therapy literature, depression is seen as a factor  blocking systemic change (Hoffman, 1981), partly by reducing problem- solving ability and partly by replacing unstable symmetrical relationships 

with stable "one-up/one-down" relationships in which the one-down position  is associated with depressed mood (Haley, 1963).

 

Can negative feedback loops be identified?

 

Can we identify a negative feedback system maintaining homeostasis of mood?   Let us use the analogy of body temperature, which is homeostatically  maintained at a certain setting by negative feedback such as sweating if  the temperature rises and shivering if the temperature falls. If mood is  the equivalent of temperature, what are the equivalents of sweating and  shivering?  The most likely candidates are aversive and rewarding stimuli,  since these are known to lower and raise mood respectively. Where do these  stimuli come from?  The most likely sources are the people in the  individual's immediate environment. If we are dealing with stimuli from  people, we can talk about signals rather than stimuli, and postulate  "putting-down" signals for lowering mood, and "boosting" signals for  raising mood.

   What is then being regulated is not actually mood but "apparent mood",  or the mood of the individual as it appears to those around him. Of the  various manifestations of mood, it is the degree of social control  exercised by the individual which most affects fellow group members;  how  much does he allow others to control his actions, and how much does he  attempt to control the actions of others?  We must assume that the social  group allocates to each member an "exercise of control" setting which is  the level of control exercised by that individual with which the group  feels comfortable, based, probably, on that individual's prestige in the  group. If the individual appears to exercise too little control, the group  increases its boosting signals and reduces its putting-down signals to that  individual. If the individual appears to exercise too much control, he is 

perceived as "too big for his boots" and the group stops boosting that  individual and starts putting him down.

   For simplicity, let us switch from the group to the dyad, and to that  particular dyad in which there is a one-up and a one-down member (Haley,  1963), such as a marital partnership in which the husband is one-up on the  wife. The one-up member defines the relationship, including the "exercise  of control" setting of the one-down member. The husband then boosts his  wife, or puts her down, so as to keep her apparent exercise of control  identical to the "exercise of control" setting which he has allocated to  her. This setting is ideally such as to ensure that the wife's mood is  high enough to carry out her duties, but not sufficiently high to threaten  his one-upness (it is our clinical experience that such fine tuning  requires more social skill than many spouses possess).

   What seems more likely, from what we know clinically, is that the "one- up" husband (or wife) tries to keep his spouse's exercise of control a  constant amount below his own exercise of control. What is maintained  homeostatically is not the absolute level of control but the difference in  control between husband and wife, what might be called the "control gap".

   More generally, the "one-up" spouse maintains a gap on what Birtchnell  (1987) has called the "vertical dimension" which describes a number of  correlated variables such as mood, rank, self-esteem, self-confidence,  dominance, and, in the last resort, the capacity to define the relationship  rather than accept the definition provided by the other. Colloquially, we  might say he tries to maintain a constant level of "one-upness";  more  technically he tries to maintain a constant vertical-gap setting between  what he feels to be his own postion on the vertical dimension and what he  perceives his wife's to be. This "gap" model has the advantage of 

embracing the phenomena of redirected aggression;  if the husband's mood is  lowered after receiving punishment from his boss at work, he restores the 

vertical gap at home by putting his wife down (or omitting to boost her). The feedback loop is probably below conscious awareness:  even though he  may be aware that he is putting his wife down, he does not understand why  he is doing it;  and many signals intended as boosting signals are received  as putting-down signals, especially in the case of "constructive" criticism  (MacLean, 1976).

   This model is the only homeostatic model I can find which could account  for the static properties of many depressions. It is clearly  oversimplistic, speculative and based on unsystematic clinical observation. I have presented it to show that it is possible to match the obvious  positive feedback loops governing mood with at least a possible negative  feedback system, suggesting that the static depressions may represent true  homeostasis rather than just buffering or other mechanisms for preventing  change (Hogan, 1980). It is noteworthy that the mood of one individual is  determined by a "setting" controlled by another. Returning to the analogy  of body temperature, it is as if a nurse were keeping her patient's  temperature at a constant amount below her own, sponging him when the  difference became too small, piling on blankets when it became too great.

 

  A RESOLUTION OF THE PARADOX SUGGESTED FROM THE PERSPECTIVE OF FUNCTION

 

Some depressions seem to be characterised by change in the  patient/environment system, others by homeostasis. Does this mean there  are two types of depression, or can one depressive episode have both change  and homeostatic functions, either at the same time or consecutively?  Are  there different levels of function, as in the case of temperature control;   where, for example, a change in the activity of sweat glands can mediate  homeostasis of body temperature?

   One possible solution can be discerned if we develop Haley's idea, 

mentioned above, that depression is associated with the one-down position  in an important relationship. If we go a little further and say that the  function of depression is to reconcile the individual to the one-down  position, then we can see that this reconciliation may in some cases be  concerned with homeostasis and at other times with change. If the  individual was formerly in the one-down position, all that is required is  that he or she should be reconciled to remain in that position in spite of  whatever new circumstances might be tending to promote change. If, on the  other hand, the individual was formerly in the one-up position in the  relationship, there must be change in the form of a reversal of dominance  in the relationship. In that case we might expect the depression to show  change characteristics while the reversal was taking place, and then  possibly to stabilise into a homeostatic depression if further  reconciliation to the one-down position were required.

   This solution accounts for the similarities of all depressions, in that  they all have the same function of reconciliation to the one-down position;   and it accounts for some of the differences between depressions, in that  there are two different starting-off points. This will be considered  further in the section on the classification of depression. The mental  state of depressed patients tends to be characterised by what Gilbert  (1989) has called "involuntary subordinate self perception", and might have  been specifically designed by evolution for the purpose of accommodating  them to the subordinate role, and other features such as apathy,  hopelessness and anxiety might have been designed to inhibit any attempts  to escape from it and gain or regain the one-up position.

 

The end-point and other changes

 

In the last section I suggested that the end-point of a "change" depression  is a reversal of dominance in the relationship. One of the concomitants of  this is passivity and compliance on the part of the newly depressed  individual. This may or may not lead to secondary change in matters which  were in conflict. Say, for instance, that a subordinate wife wants to move  house, but her dominant husband refuses. After some marital conflict, the  dominance relations are reversed, the husband becomes depressed, and the  wife gets her own way. The move of house is secondary change, dependent on  the reversal of dominance. However, the move of house may appear to be the  end-point, as this was the wife's objective in the struggle, and when it is  achieved she no longer nags the husband. If it is the husband who wants to  move house, then the end-point for the wife is not "moving house" but "not  moving house", in other words, the absence of secondary change (except in  the form of a change in aspiration on the part of the husband, in that he  gives up the idea of moving house). The primary change is not the move of  house, but who decides whether or not they move house.

   Another form of secondary change may be part of the positive feedback  process of the "change" depression. Suppose in the above example that the  husband loses his job. This change was not the end-point desired by the  wife, but it deprives the husband of standing in the family and facilitates  the reversal of dominance, leading eventually to the desired change.

                   OBJECTIONS TO THE HYPOTHESIS

 

Why are some subordinates not depressed?

 

 

Depression is only one of a number of mechanisms for facilitating  subordination during the course of dyadic interaction. Therefore all  subordinate individuals are not depressed. Children tend to be subordinate  to their parents without the need for depression, and in general if there  is a great difference in size or age or skill between two individuals, one  may automatically adopt a subordinate role without any suggestion of  depression;  in fact, there may be joyous surrender to one whom the  subordinate individual respects or even adulates. Depression is likely to  be a "safety net" mechanism to ensure asymmetry when other methods of  negotiation have failed. It is likely to occur when rank is contested, or  when extra demands are made on the subordinate member - demands which in  the absence of depression might incite them to rebellion.

 

Why does depression occur after loss?

 

The dyadic interaction of agonistic behaviour is the main vertebrate  mechanism for inducing social asymmetry, but in human beings it has been  superceded to a large extent by other forms of social competition. In  human social life, asymmetry is often imposed on members of a dyad from  outside, in two rather different ways. Dominant rank may depend on a  patron who favours one member at the expense of the other, so that rank  reversal may occur if a patron is lost. This may lead to a "change"  depression in the member who loses rank;  whereas if the subordinate member  loses a patron the subordinate rank would be intensified and a "static"  depression could be expected. Also, as Gilbert (1989) has pointed out,  instead of displaying their power to each other in an attempt to intimidate  the other, two rivals may present themselves as attractive to the group as  a whole in an attempt to solicit approbation, which may lead to differences  of prestige and self-esteem. Possibly this evolutionarily new form of 

social competition has been built on the mechanisms of dyadic agonistic  behaviour, so that a change depression may occur in someone who loses the  approbation of the group, whereas a static depression occurs in someone who  has never had the approbation of the group. In human social life,  subordination is induced by agonistic behaviour only in settings in which  society has neither the will nor the power to intervene, such as the  street-corner gang, the school playground and the nuclear family;   therefore it is only in these settings that we are likely to see depression  resulting directly from dyadic interaction.

 

Which relationship?

 

To the suggestion that the function of depression is concerned with the  issue of being one-down in the relationship, it might be argued that human  beings have many relationships, so how do we know which one is liable to  generate the kind of dyadic interaction which might lead to depression?

   Clinically, we find that a depressed person is usually having problems  with one relationship, and it may be a spouse, a parent, a child, a  sibling, an employer, or some other person who is important to the patient.

   Over hundreds of millions of years of our evolution, we probably lived  in groups with a fairly linear dominance hierarchy, like present day  baboons and macaques (Barkow, 1975) in which all dyadic relationships are  complementary in terms of power. In these monkey societies there are only  two relationships that are likely to be contested, one with whoever ranks  below and the other with whoever ranks above. If there is difficulty with  the monkey who ranks below, this becomes the priority relationship, and  that with the monkey above is likely to be put on "hold" (in the form of  submission) while the first is being sorted out. Therefore our minds have  evolved to "take on" only one rival at a time.

 

 

 

        IMPLICATIONS FOR THE CLASSIFICATION OF DEPRESSION

 

How does this hypothesis relate to the idea of two types of depression?   The depression of a one-up person becoming a one-down person is likely to  be more severe than that of a one-down person remaining one-down, and  onlookers are more likely to comment on change in behaviour or attitude,  but since the objective of the depression is the same in the two cases  (adjustment to the one-down position) one would expect the depressions to  be quite similar. This fits with the distinction between "neurotic" and  "endogenous" depression (Price, 1969). The difference between the two  depressions is a difference in starting-point rather than of content.

   Beck's (1976) negative cognitive triad (about the self, the world and  the future) is appropriate to both types of depression, in that such  thinking disposes the sufferer to accept whatever has been imposed on him  without attempting rebellion. But a negative view of the past is  appropriate to a change depression, in which thinking should take such  forms as "my former rank was inappropriate" and "my former successes were  all a sham", with which, indeed, the clinician is familiar in his  psychotically depressed patients. Such thinking need not occur in the  static depressions because there never has been any previous success or  high rank to form the subject of cognitive distortion.

   Possibly the view of neurotic and endogenous depressions as having  homeostatic and change functions, respectively, may help researchers to  separate the two clinically. For instance, rating scales attempting to  separate the two depressions might well concentrate on the statements of  informants, who notice a change from previous personality in patients with  endogenous depression, but concerning patients with neurotic depression  tend towards the view that "he is much the same as he has always been, only 

rather more so".

 

 

                  IMPLICATIONS FOR RESEARCH

 

Evolutionary hypotheses are not directly testable, and so it is important  for them to be heuristic and to provide a novel conceptual scheme from  which refutable hypotheses can be generated. This applies to the idea that  depression evolved because it served the function of enabling our ancestors  to adopt subordinate social roles. The possible homology between human  depression and animal defeat reactions offers some promising animal models  (Henry, 1982; Leshner, 1983; McGuire, 1988; Sapolsky, 1989; Price, 1989).

   In human beings, the hypothesis orients us to the psychology of  complementary relationships. Bateson (1972), Sluzki and Beavin (1977),  Hinde (1979) and the "interpersonal" psychologists (Orford, 1986) have  given serious consideration to the topic, but more research is needed. We  need to develop measuring instruments which will give reliable measures of  consistent asymmetry in relationships, so far not achieved (Gray-Little &  Burks, 1983).

   With instruments to measure complementarity and the expression of both  boosting and putting-down signals between the members of a dyad, we would  be in a position to make predictions from the theory:  for instance that,  in complementary marriages, the onset of depression in the one-up partner  (but not in the one-down partner) would be associated with an increase in  aversive signalling (expressed hostility) to the spouse. A preliminary  study gave some evidence for this (Price, 1988) but a far more rigorous  study is required. Previous work on the expression of hostility in  depression has not taken account of complementarity between the patient   and the object of the hostility, and it is possible that the neglect of  this variable has been responsible for the conflicting findings which have  resulted from such studies (Riley et al., 1989).

 

 

                      RELATION TO OTHER THEORIES

 

The "change" depressions postulated here are consistent with the  "disengagement from incentives" theory of depression put forward by Klinger  (1975), in that they function to disengage the patient from a desirable  social role which is no longer tenable. The "static" depressions are  consistent with the learned helplessness theory of depression put forward  by Seligman (1975), in that they discourage the patient from resisting the  aversive stimulation which may pass down a social hierarchy, and thus serve  the function of maintaining the stability of the complementary  relationships which constitute a social hierarchy.

   We can also attempt to integrate this phylogenetic view with  interpersonal psychology. Birtchnell (1987) has called the vertical  dimension of the interpersonal circle directiveness/receptiveness or  upperness/lowerness. According to our theory, a "change" depression serves  the function of moving an individual down the vertical dimension from  upperness to lowerness;  a "static" depression serves to maintain an  individual in a position of lowerness. The horizontal dimension of  closeness/distance is clearly uncorrelated with these functions, as one can  have one's place usurped as readily by a stranger as by one's brother. However, at least one other dimension is required to fit the model with  reality, as we know that states of lowerness may be either welcome or  unwelcome, and this dimension is important for the affective states of the  actors concerned. A position of lowerness may be adopted willingly, even  joyfully, when the position of upperness is occupied by someone who is  respected or loved, and from whom security and praise may be forthcoming;   this position of lowerness is complementary to the "idealised other" of  Kohut or the "hero archetype" of Jung (Gilbert, 1991). On the other hand, 

if the upper individual is resented rather than respected, the lower person 

has been coerced into lowerness and is likely to be tempted to rebellion;   it is in preventing this rebellion that depression serves its function.

   This variation between joyful and resentful lowerness is, I think, best  expressed by Chance's concept of hedonic and agonic modes (Chance, 1988),  because in the hedonic mode status relationships are not an issue whereas  in the agonic mode the actors are oriented towards contesting status  relationships with agonistic behaviour. In describing relationships, we  may think of upperness/lowerness (or, more correctly, symmetry/com-  plementarity) and closeness/distance as trait variables whereas  hedonic/agonic is a state variable which reflects the degree to which the  actors are currently satisfied with the existing status relationships. Although some loving or hero-worshipping relationships may be very stable  over long periods, others may switch rapidly from hedonic to agonic and  back again in processes of dissuagement (Heard and Lake, 1986) and  reconciliation (de Waal, 1989), and this is typically seen in marital  relationships. There are two exit routes from agonic lowerness  (depression):  one climbs towards upperness (or symmetry), the other is a  switch to hedonic lowerness (implying that the lowerness and its  implications are accepted)..

 

                        CONCLUSIONS

 

The theoretical arguments presented here support the long-held view that  there are two distinct types of human depression. One type serves a static  function;  namely, to maintain the one-down member of a relationship in  that position in spite of motivation to become one-up and in spite of  demanding and provocative behaviour by the one-up member. The other type  serves a change function;  namely, to facilitate the change of the one-up  member in a relationship into the one-down position, in spite of motivation 

to remain one-up. To some extent, both types of depression serve the same  function;  namely, to reconcile the one-down member of a relationship to  what is likely to be an uncomfortable and unrewarding social role. They  differ in their point of origin, in that one is helping the actor to remain  where he has been, whereas the other is helping him to change from his  former one-up position.

   This hypothesis is consistent with ethologically-based theories which  see depression as an evolved mechanism concerned with the maintenance of  asymmetry in relationships. The capacity to live harmoniously in equal,  symmetrical, reciprocal relationships with others is rare among  vertebrates. In almost all species, two same-sexed adults sharing a  territory develop an asymmetrical relationship, whereby one is dominant and  the other subordinate. The subordinate animal needs to lead a life of  considerable inhibition in many activities, such as feeding, mating,  exploration and freedom of movement;  in fact, it needs to have a different  mentality or behavioural style from the dominant animal. We could call  this the subordinate mentality. Because animal social life depends for its  success on the development and maintenance of a subordinate mentality in a  considerable proportion of any population, it is likely that several  mechanisms have evolved for inducing subordination. It has been suggested  by a number of writers that human depression is a manifestation of one of  these subordination-inducing mechanisms (Price, 1967; Gardner, 1982 and  1988; Hartung, 1987; Gilbert, 1989 and 1991; Sloman et al., 1989). The  depression is part of a system of dyadic interaction which ensures the  development and maintenance of asymmetry.

   Regarding prophylaxis and therapy, depression is seen as a primitive,  involuntary means of resolving interpersonal conflict. The necessity for 

it can be avoided if negotiation can produce a compromise solution based on  reciprocity and interpersonal equality;  or, if yielding by one party is 

unavoidable, voluntary yielding can replace the involuntary yielding of  depression. In this sense, the experience of depression contains the seeds  of its own resolution by inducing a state of "giving in and giving up". The therapist can facilitate this process by assisting the patient to give  up unequal struggles, unrealisable goals and unachievable aspirations.

   Human social life is very different from the social organisation in  which depressive states are likely to have evolved. The social hierarchies  based on intimidation have given way largely to status systems based on the  display of attractiveness and the voluntary conferral of power. Even  beyond this, for many people the experience of winning or losing has become  an inner symbolic one, detached from the realities of the social situation  they are in. As Longfellow said:

 

                Not in the clamour of the crowded street,

                Not in the shouts and plaudits of the throng,

                But in ourselves, are triumph and defeat.

 

However, nature is a tinkerer rather than an engineer, and it is likely  that the mechnisms subserving these advanced forms of triumph and defeat  have been built onto the foundations of the old ones, so that they may  still trigger affective states which were functional in relation to the  primitive hierarchies. And these primitive hierarchies may even now be  discerned in places where cultural influences are not pronounced, such as  street gangs, school playgrounds and the matrimonial home. In these  situations we can actually see depression facilitating reversal of rank or  the maintenance of low rank, and in such cases the systemic properties of  change and stasis are of functional importance.

 

                         References

 

 

Barkow, J.H. (1975) Prestige and culture: a biosocial interpretation.

  Current Anthropology, 16, 553-572.

Bateson, G. (1972) Steps to an Ecology of Mind. New York: Ballantine Books.

Beck, A.T. (1976) Cognitive Therapy and the Emotional Disorders. New York:

  International Universities Press.

Birtchnell, J. (1987) Attachment-detachment, directiveness-receptiveness: a

  system for classifying interpersonal attitudes and behaviour. British

  Journal of Medical Psychology, 60, 17-27.

Brewin, C.R. (1988) Cognitive Foundations of Clinical Psychology. Hove: 

  Lawrence Erlbaum.

Brown, G.W. (1989) Depression:  a radical social perspective. In

  K.R.Herbst & E.S.Paykel (Eds) Depression: an Integrative Approach, pp.

  21-44. Oxford: Heinemann.

Chance, M.R.A. (1988) Introduction. In M.R.A.Chance (Ed) Social Fabrics of

  the Mind pp. 1-35. Hove: Lawrence Erlbaum.

Costello, C.G. (1976) Anxiety and Depression:  The Adaptive Emotions.

  Montreal:  McGill-Queen's University Press.

de Waal, F. (1989) Peacemaking Among Primates. Cambridge: Harvard

  University Press.

Feldman, L.B. (1976) Depression and marital interaction. Family Process,

  15, 389-396.

Gardner, R.J. Jr. (1982) Mechanisms in major depressive disorder: an

  evolutionary model. Archives of General Psychiatry, 39, 1436-1441.

Gardner, R.J. Jr. (1988) Psychiatric syndromes as infrastructure for

  intraspecific communication. In Social Fabrics of the Mind (ed.

  M.R.A.Chance) Hove: Lawrence Erlbaum.

Gilbert, P. (1989) Human Nature and Suffering. Hove: Lawrence Erlbaum.

Gilbert, P. (1991) Depression: Types, Concepts and Theories; an

  evolutionary synthesis on the themes of power and belonging. Hove:

 

  Lawrence Erlbaum (in press).

Gray W. (1969)  History and development of General Systems Theory. In

  W.Gray, F.J.Duhl and N.D.Rizzo (Eds) General Systems Theory and

  Psychiatry, pp. 7-31. Boston:  Little, Brown & Co.

Gray-Little, B. & Burks, N. (1983) Power and satisfaction in marriage: a

  review and critique. Psychological Bulletin, 93, 513-538.

Haley, J. (1963) Marriage therapy. Archives of General Psychiatry, 8, 213-

   234.

Hartung, J. (1987) Deceiving down: Conjectures on the management of

  subordinate status. In J. Lockard and D Pulhus (Eds) Self-deceit: an

  Adaptive Strategy. Englewood Cliffs, NJ: Prentice-Hall.

Heard, D.H. & Lake, B. (1986) The attachment dynamic in adult life. British

  Journal of Psychiatry, 149, 430-438.

Henderson, A.S. (1974)  Care-eliciting behavior in man. Journal of Nervous

  and Mental Diseases, 159, 172-181.

Henry J.P. (1982)  The relation of social to biological processes in

  disease. Social Science and Medicine, 16, 369-380.

Hinde, R.A. (1979) Towards Understanding Relationships. London: Academic.

Hoffmann, L. (1981) The Foundations of Family Therapy. New York: Basic

  Books.

Hogan, J.A. (1980) Homeostasis and behaviour. In F.M.Toates and T.R.

  Halliday (Eds) Analysis of Motivational Processes, pp. 3-21. London:

  Academic Press.

Klerman, G.L. (1974) Depression and adaptation. In R.J.Friedman and

  M.M.Katz (Eds) The Psychology of Depression, pp. 127-145. Washington,

  D.C.:  V.H.Winston.

Klinger, E. (1975) Consequences of commitment to and disengagement from

  incentives. Psychological Review, 82, 1-25.

Leshner, A.I. (1983) The hormonal responses to competition and their

 

  behavioral significance. In B.B.Svare (Ed.) Hormones and Aggressive

  Behavior, pp.     New York:  Plenum Press.

McGuire, M.T. (1988) On the possibility of ethological explanations of

  psychiatric disorders. Acta Psychatrica Scandinavica, 77, Supp.341, 7-22.

MacLean, P.D. (1976) Depression as a specific response to stress. In

  I.G.Sarason and C.D.Spielberger (Eds) Stress and Anxiety, pp. 297-323.

  New York:  John Wiley.

Marmor, J. (1983) Systems thinking in psychiatry: some theoretical and

  clinical implications. American Journal of Psychiatry, 140, 833-838.

Maruyama, M. (1963) The second cybernetics: deviation-amplifying mutual

  causal processes. American Scientist, 51, 164-179.

Orford, J. (1986) The rules of interpersonal complementarity: does

  hostility beget hostility and dominance, submission? Psychological

  Review, 93, 365-377.

Paykel, E.S. (1978) Contribution of life events to the causation of

  psychiatric illness. Psychological Medicine, 8, 245-253.

Price, J.S. (1967) Hypothesis: the dominance hierarchy and the evolution of

  mental illness. Lancet, 2, 243-246.

Price, J.S. (1969) Neurotic and endogenous depression:  a phylogenetic

  view. British Journal of Psychiatry, 114, 119-120.

Price, J.S. (1988) Alternative channels for negotiating asymmetry in social

  relationships. In M.R.A.Chance (Ed) Social Fabrics of the Mind, pp. 157-

  195. Hove: Lawrence Erlbaum.

Price, J.S. (1989) The effect of social stress on the behaviour and

  physiology of monkeys. In K. Davison & A. Kerr (Eds) Contemporary

  Themes in Psychiatry, pp. 459-466. London: Gaskell.

Price, J.S. & Sloman, L. (1987) Depression as yielding behavior:  an

   animal model based on Schjelderup-Ebbe's pecking order. Ethology and

   Sociobiology, 8, 85-98 (Supplement).

 

Pyszczynski, T. & Greenberg, J. (1987) Depression, self-focused attention

  and self-regulatory perseveration. In C.R.Snyder and C.E.Ford (Eds)

  Coping with Negative Life Events, pp. 105-129. New York: Plenum.

Riley, W.T., Treiber, F.A., Woods, M.G. (1989) Anger and hostility in

  depression. Journal of Nervous and Mental Disease, 177, 668-674.

Sapolsky, R.M. (1989) Hypercortisolism among socially subordinate wild

  baboons originates at the CNS level. Archives of General Psychiatry, ,

  1047-1051.

Seligman, M.E.P. (1975) Helplessness. San Fransisco:  Freeman.

Senay E.C. (1973)  General systems theory and depression. In P.J.Scott and 

  E.C.Senay (Eds) Separation and Depression., pp. 237-245. Washington,

  D.C.:  American Association for the Advancement of Science.

Sloman, L., Berridge, M., Homatidis, M.A., Hunter, D. and Duck, T. (1982).

   Gait patterns of depressed patients and normal subjects. American

  Journal of Psychiatry, 139, 94-97.

Sloman, L., Gardner, R. & Price, J.S. (1989) Biology of family systems and

  mood disorders. Family Process, 28, 387-398.

Sluzki, C.E. and Beavin, J. (1977) Symmetry and complementarity: an

  operational definition and a typology of dyads. In P. Watzlawick & J.H.

  Weakland (Eds) The Interactional View, pp. 71-87. New York: W.W.Norton.

Teasdale, J.D. (1985) Psychological treatments for depression:  how do they

  work?  Behaviour Research and Therapy, 23, 157-165.

Veiel, H.O.F. & Kuhner, C. (1990) Relatives and depressive relapse: the

  critical period after dischange from in-patient treatment. Psychological

  Medicine, 20, 977-984.

Wender, P.H. (1968) Vicious and virtuous circles: the role of deviation

  amplifying feedback in the origin and perpetuation of behavior.

  Psychiatry, 31, 309-324.